Re: excessive kworker activity when idle. (was Re: vma corruption intoday's -git)

[Linus Torvalds]

On Wed, Mar 30, 2011 at 8:55 PM, Dave Jones <davej@xxxxxxxxxx> wrote:
>
> I tried that, and wasn't particularly enlightened.
>
> +      6.53%  kworker/1:2  [kernel.kallsyms]  [k] read_hpet
> +      4.83%  kworker/0:0  [kernel.kallsyms]  [k] read_hpet

Go into those, deeper into the callchain. That's how to see what the
upper-level problem is - by seeing who actually calls it. The leafs
themselves (where the time is obviously spent) don't tend to be very
informative, no.

>  > (Powertop can also do it, and is probably a better thing to use, I'm
>  > just used to "perf record" for other reasons, so..)
>
> Tried that too, here's what it said..

Hmm. Powertop is _supposed_ to be able to look into the kwork things
(it has special trace points for it), but as mentioned, the times I've
seen it I've used 'perf', so I haven't actually tried it myself.

                  Linus
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/